DESCRIPTION: Nonactin is an inhibitor of the 170-kDa-P-glycoprotein mediated efflux of 4'-O-tetrahydropyranyladriamycin in multi-drug resistant erythroleukemia K562 cells. Nonactin and its macrotetrolide homologues are als active against cancer cell lines in vitro and have shown tumor reducing activity in in vivo studies with mice. Further, the macrotetrolides act as typical ionophore antibiotics being effective against Gram+ bacteria, fungi, and mycobacteria. Nonactin is not a current drug candidate because of its relatively high toxicity. The naturally occurring homologues, however, have greater activity than nonactin: from a QSAR analysis we propose a series of ne compounds which have potentially much higher activity. Chemical synthesis of macrotetrolide homologues is not trivial as it requires the separate synthesis of both enantiomers of nonactic acid; selective assembl of nonactic acid units into a tetrameric intermediate; and finally a macrolactonization reaction to form the macrotetrolide ring. Streptomyces griseus strain ETH A7796, in comparison, can readily make over 1 gL-1 of nonactin in fermentation. We propose to combine the best aspects of chemical synthesis and 'biosynthesis' to make new macrotetrolides. Chemical synthesis will be used to make racemic nonactic acid analogs via established, efficient routes. 'Biosynthesis' in a genetically altered strain of S. griseus will achieve the stereospecific biotransformation of the nonactic acid analogs into new macrotetrolides. The specific aims of this proposal are: 1. To purify from S. griseus ETH A7796 and characterize the key enzyme in nonactin biosynthesis, nonactate synthase, which catalyzes the conversion of a acyclic precursor into nonactic acid. 2. To isolate and characterize genes of the nonactin biosynthesis cluster. 3. To use a mutant strain, disrupted in nonactin biosynthesis, to make four prototypical macrotetrolide analogs, and to study their physical and biologica properties.